These cases illustrate the value of minute and careful sub-division of the hulls of vessels by watertight bulkheads. Unless, however, the bulkheads are carried a few feet higher than the level of the water outside—and it is to be regretted that this is still not infrequently overlooked or neglected in merchant steamers—they are valueless, and, indeed, had better not be in the ship at all. They will contribute to the loss of the vessel by keeping the water at one end, and carrying her bows under, whereas if they are not fitted, the same volume of water will distribute itself throughout the bottom of the ship fore and aft, preserve the even trim of the vessel, and allow more pumps to cope with the inflow. Although her freeboard, or height of side above water will be reduced, she will still be seaworthy, the boiler fires may be kept burning, and the machinery going, sufficiently long for her to reach a port of safety. Readers appreciating the above considerations will readily see why it is that sailing vessels are usually fitted with only one transverse bulkhead—that near the bow—and understand how it is that the outcry sometimes made by inexperienced people about the absence of other bulkheads in emigrant sailing vessels is for most part unheeded by those on whom the responsibility falls. From statistics presented in the paper above referred to, it is shown that during a period of six years, ending with December, 1882, the average loss per annum of ships not qualified for the Admiralty list was one in twenty-five; while of ships so qualified the annual average loss was only one in eighty-six. The chances of loss from any cause are thus seen to be nearly four times as great for a ship not constructed to qualify for the Admiralty list as for a vessel entered on that list. During the first four-and-a-half years of the period referred to, not one ship of those entered on the list was lost by collision although a considerable number had been in collision, and escaped foundering by reason of the safety afforded by their bulkheads. During 1882 six casualties happened to ships on the list, one of which—a case of collision—proved fatal. This was a case, however, such as no merchant steamer afloat at the time would have been capable of surviving. The whole of the ship—a small one—was flooded abaft the engine-room, the two after holds being open to the sea. The whole of the losses from the Admiralty list during the period referred to—eleven in number—have been from drifting on rocks, or otherwise getting fixed on shore, with the solitary exception above quoted. In the same period 76 ships have been lost which had been offered for admission to the list, but had not been found qualified; of these 17, or 22½ per cent., were lost by collision; and 10, or 13¼ per cent., were lost by foundering; most of the rest stranded or broke up on rocks. The risk of fatal collision, according to Mr Dunn, is about 1 to 100, irrespective of the class of ship, and